Cai Yiyang - Home page

👋 Hi there! I am CAI Yiyang (蔡逸扬), a Ph.D. student of Hong Kong University of Science and Technology. I am supervised by Prof. GUO Yike and Prof. LUO Wenhan. My research interests include computer vision and generative models. My main focusing topic is about ID-preserving content generation.
Prior to my doctoral studies, I worked as a machine learning engineer at Intel for two years, where I worked on model optimization and acceleration techniques about diffusion models and LLMs.
I hold a Master’s degree from the University of California, Berkeley and a Bachelor’s degree from Beijing University of Aeronautics and Astronautics (Beihang University).


đź’Ľ Education & Working Experience

2024.08 - Present

Ph.D. student of Independent Interdisciplinary Program

2022.06 - 2024.06

Machine learning engineer of Data Center and Artificial Intelligence (DCAI) Group

2020.08 - 2021.12

Graduate student of Electrical Engineering and Computer Science Department

2020.09 - 2021.11

Algorithm intern, working with Dr. LIU Jiaming.

2016.08 - 2020.07

Undergraduate student of Shenyuan Honor College


📝 Publications

Foundation Cures Personalization: Improving Personalized Models' Prompt Consistency via Hidden Foundation Knowledge

Yiyang Cai, Zhengkai Jiang, Yulong Liu, Chunyang Jiang, Wei Xue, Wenhan Luo, Yike Guo
Technical Report, 2025
[paper] [code] [page]

Optimize Weight Rounding via Signed Gradient Descent for the Quantization of LLMs

Wenhua Cheng, Weiwei Zhang, Haihao Shen, Yiyang Cai, Xin He, Kaokao Lv, Yi Liu
Findings of EMNLP, 2024
[paper] [code]

Effective Quantization for Diffusion Models on CPUs

Hanwen Chang, Haihao Shen, Yiyang Cai, Xinyu Ye, Zhenzhong Xu, Wenhua Cheng, Kaokao Lv, et al.
NeurIPS 2023 Workshop on Diffusion Models.
[paper]

TEQ: Trainable Equivalent Transformation for Quantization of LLMs

Wenhua Cheng, Yiyang Cai, Kaokao Lv, Haihao Shen
Technical Report, 2023
[paper] [code]

Few-shot Font Generation by Learning Fine-grained Local Styles

Licheng Tang*, Yiyang Cai*, Jiaming Liu, Zhibin Hong, Mingming Gong, Minhu Fan, Junyu Han, et al. ('*' means equal contribution)
CVPR, 2022
[paper] [code]